What Time Is It? How Data Geometry Makes Time Conditioning Optional for Flow Matching

• URL: http://arxiv.org/abs/2605.08344v1
• Date: Fri, 08 May 2026 18:00:36 GMT
• Title: What Time Is It? How Data Geometry Makes Time Conditioning Optional for Flow Matching
• Abstract summary: Flow matching models can be trained without explicit time conditioning, challenging the standard view that the time is needed to disambiguate velocity targets.<n>Decomposing the time-blind flow matching loss, we identify two sources of irreducible error.<n>We show that the geometry of high-dimensional data makes time identifiable directly from noisy observations.
• Score: 30.271879683388033
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Recent work has shown that models flow matching models can be trained without explicit time conditioning, challenging the standard view that the interpolation time is needed to disambiguate velocity targets. But why should a time-blind model work at all? Decomposing the time-blind flow matching loss, we identify two sources of irreducible error: a coupling variance, which arises from ambiguous velocity targets induced by how noise and data points are paired, and the time-blindness gap, which is the additional error caused by ignoring time. This gap shows that time-blind training is strictly harder than conventional training, reinforcing the puzzle that time-blind models work so well in practice. We resolve this tension by showing that the geometry of high-dimensional data makes time identifiable directly from noisy observations. When data concentrates near a $k$-dimensional subspace, time can be recovered from the statistical structure of noisy interpolants in directions orthogonal to the data; under a spiked-covariance model, this yields a closed-form estimator that recovers $t$ from a single observation $z$ at rate $O(1/\sqrt{d-k})$ for ambient dimension $d$. As a consequence, we prove that the time-blindness gap is asymptotically negligible relative to the coupling variance. We empirically demonstrate our identifiability result on real-world data and show that changing the coupling has a much larger effect on loss and sample quality than removing time conditioning across CIFAR-10, CelebA-HQ, and FFHQ. These results explain why time-blind flow matching works and show that the main practical lever is the choice of coupling, not explicit time conditioning.

Related papers

• A theoretical framework for self-supervised contrastive learning for continuous dependent data [79.62732169706054]
  Self-supervised learning (SSL) has emerged as a powerful approach to learning representations, particularly in the field of computer vision.<n>We propose a novel theoretical framework for contrastive SSL tailored to emphsemantic independence between samples.<n>Specifically, we outperform TS2Vec on the standard UEA and UCR benchmarks, with accuracy improvements of $4.17$% and $2.08$%, respectively.
  arXiv  Detail & Related papers  (2025-06-11T14:23:47Z)
• T-Graphormer: Using Transformers for Spatiotemporal Forecasting [2.855856661274715]
  T-Graphormer reduces root mean squared error (RMSE) and mean absolute percentage error (MAPE) by up to 20% and 10%.<n>We show the effectiveness of T-Graphormer on real-world traffic prediction benchmark datasets.
  arXiv  Detail & Related papers  (2025-01-22T23:32:29Z)
• SpaceTime: Causal Discovery from Non-Stationary Time Series [34.39247638413985]
  Understanding causality is challenging and often complicated by changing causal relationships over time and across environments.<n>Existing methods for discovering causal graphs from time series either assume stationarity, do not permit both temporal and spatial distribution changes, or are unaware of locations with the same causal relationships.<n>In this work, we unify the three tasks of causal graph discovery in the non-stationary multi-context setting.<n>We show that our method provides insights into real-world phenomena such as river-runoff measured at different catchments and biosphere-atmosphere interactions across ecosystems.
  arXiv  Detail & Related papers  (2025-01-17T15:00:20Z)
• Graph Spatiotemporal Process for Multivariate Time Series Anomaly Detection with Missing Values [67.76168547245237]
  We introduce a novel framework called GST-Pro, which utilizes a graphtemporal process and anomaly scorer to detect anomalies. Our experimental results show that the GST-Pro method can effectively detect anomalies in time series data and outperforms state-of-the-art methods.
  arXiv  Detail & Related papers  (2024-01-11T10:10:16Z)
• LARA: A Light and Anti-overfitting Retraining Approach for Unsupervised Time Series Anomaly Detection [49.52429991848581]
  We propose a Light and Anti-overfitting Retraining Approach (LARA) for deep variational auto-encoder based time series anomaly detection methods (VAEs) This work aims to make three novel contributions: 1) the retraining process is formulated as a convex problem and can converge at a fast rate as well as prevent overfitting; 2) designing a ruminate block, which leverages the historical data without the need to store them; and 3) mathematically proving that when fine-tuning the latent vector and reconstructed data, the linear formations can achieve the least adjusting errors between the ground truths and the fine-tuned ones.
  arXiv  Detail & Related papers  (2023-10-09T12:36:16Z)
• Temporal Smoothness Regularisers for Neural Link Predictors [8.975480841443272]
  We show that a simple method like TNTComplEx can produce significantly more accurate results than state-of-the-art methods. We also evaluate the impact of a wide range of temporal smoothing regularisers on two state-of-the-art temporal link prediction models.
  arXiv  Detail & Related papers  (2023-09-16T16:52:49Z)
• Imbalanced Aircraft Data Anomaly Detection [103.01418862972564]
  Anomaly detection in temporal data from sensors under aviation scenarios is a practical but challenging task. We propose a Graphical Temporal Data Analysis framework. It consists three modules, named Series-to-Image (S2I), Cluster-based Resampling Approach using Euclidean Distance (CRD) and Variance-Based Loss (VBL)
  arXiv  Detail & Related papers  (2023-05-17T09:37:07Z)
• INCREASE: Inductive Graph Representation Learning for Spatio-Temporal Kriging [35.193756288996944]
  SRU-temporal kriging is an important problem in web and social applications, such as Web or Internet of Things, where things (e.g., sensors) connected into a web often come with spatial and temporal properties. We propose a novel inductive graph representation learning model for kriging.
  arXiv  Detail & Related papers  (2023-02-06T12:28:35Z)
• Realization of Causal Representation Learning to Adjust Confounding Bias in Latent Space [28.133104562449212]
  Causal DAGs(Directed Acyclic Graphs) are usually considered in a 2D plane. In this paper, we redefine causal DAG as emphdo-DAG, in which variables' values are no longer time-stamp-dependent, and timelines can be seen as axes.
  arXiv  Detail & Related papers  (2022-11-15T23:35:15Z)
• TILDE-Q: A Transformation Invariant Loss Function for Time-Series Forecasting [8.086595073181604]
  Time-series forecasting can address real-world problems across various domains, including energy, weather, traffic, and economy. Time-series forecasting is a well-researched field, predicting complex temporal patterns such as sudden changes in sequential data still poses a challenge with current models. We propose a novel, compact loss function called TILDEQ that considers not only amplitude and phase distortions but also allows models to capture the shape of time-series sequences.
  arXiv  Detail & Related papers  (2022-10-26T21:32:20Z)
• Anomaly Transformer: Time Series Anomaly Detection with Association Discrepancy [68.86835407617778]
  Anomaly Transformer achieves state-of-the-art performance on six unsupervised time series anomaly detection benchmarks. Anomaly Transformer achieves state-of-the-art performance on six unsupervised time series anomaly detection benchmarks.
  arXiv  Detail & Related papers  (2021-10-06T10:33:55Z)
• Time-Series Imputation with Wasserstein Interpolation for Optimal Look-Ahead-Bias and Variance Tradeoff [66.59869239999459]
  In finance, imputation of missing returns may be applied prior to training a portfolio optimization model. There is an inherent trade-off between the look-ahead-bias of using the full data set for imputation and the larger variance in the imputation from using only the training data. We propose a Bayesian posterior consensus distribution which optimally controls the variance and look-ahead-bias trade-off in the imputation.
  arXiv  Detail & Related papers  (2021-02-25T09:05:35Z)
• A Case-Study on the Impact of Dynamic Time Warping in Time Series Regression [2.639737913330821]
  We show that Dynamic Time Warping (DTW) is effective in improving accuracy on a regression task when only a single wavelength is considered. When combined with k-Nearest Neighbour, DTW has the added advantage that it can reveal similarities and differences between samples at the level of the time-series. However, in the problem, we consider here data is available across a spectrum of wavelengths.
  arXiv  Detail & Related papers  (2020-10-11T15:21:21Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.